Emma Uprichard, Associate Professor, Centre for Interdisciplinary Methodologies
Published May 2015
What does your work in data science at Warwick entail?
Mostly my work is methodological. More specifically, I am interested in developing methods that describe and explain change and continuity over time and space in a way that's also useful for social policy and planning purposes. I do this by looking more closely at many things including cities, food, the lifecourse and constructs of childhood.
What are the dangers present in big data research?
Many methods that are used for exploring and analysing big data are not necessarily conducive to developing useful social policies. This is because mostly they tend to model change in variables rather than *cases* and they also assume that change is linear. However, when it comes to social policy and planning, we need a range of methodological approaches so that it is possible to explore how similar initial conditions can lead to multiple outcomes, and vice versa, how different initial conditions can lead to similar outcomes.
In other words, there are multiple pathways to particular states of being and we need methods that allow for this multiplicity. For example, it is possible to become unemployed or socially excluded through a variety of sequences of events and reasons. Likewise, it is possible to improve a ‘poorly performing’ school in a variety of ways too. The ‘messiness’ and the sheer contingency involved in shaping social life are not well dealt with in big data methodologies. Therefore, whilst big data research may be great for some things, many people working in policy planning and practice are very sceptical about how useful big data will be in terms of solving some of our most challenging social problems.
How important is it to develop big data research and skills?
Very! It’s fundamental to what is needed. Initiatives like the Nuffield/ESRC/HEFCE Q-Step Centres (e.g. Warwick Q-Step Centre) recognise that new skills need to be taught at undergraduate level, not just postgraduate level and that disciplines such as those in the social sciences need to engage immediately in skill-up for future graduates to understand, participate and intervene in today’s data-driven digital society.
In your opinion where does big data have the potential to create the biggest impact?
In society. Most big data is social data – it will and does already impact on society and social transformations and both local and global social issues.
What can data science and big data do for the average citizen?
It can be good, because it could improve health diagnostics, improve transport, improve recommendations about what to buy or do, when and where to go, et cetera. But it might also be more problematic, for example if new tools are developed that can ‘pre-predict’ what individuals might do, eat, buy, do, go - and that information is used to set individual prices, insurance and mortgage rates.
How important is wider understanding of big data and data science?
I think it's tremendously important. Big data will inform different parts of our lives in the future from healthcare, to travel, to business and engineering. It would be fantastic if we could get everyone understanding what big data is all about. That's why events such as the Cheltenham Science Festival are so important. We also have MOOCs such as Warwick's Big Data: Measuring and Predicting Human Behaviour, which are a great introduction to studying big data.